Researchers say they've developed a concept for a fusion reactor that could be built for less money than an equivalent coal-fired plant — but they acknowledge that they still have some questions to answer. For example, will the concept really work?
The design concept, known as the "dynomak," is the subject of a detailed economic analysis as well as a presentation to be made next week in Russia at the International Atomic Energy Agency's 25th Fusion Energy Conference. The analysis suggests that a dynomak capable of producing 1 gigawatt of electrical power could be built for $2.7 billion, compared with $2.8 billion for a comparable coal plant.
That's far less than the estimated $50 billion-plus price tag for the 35-nation ITER demonstration fusion reactor that's being built in France, with a target date of 2027 for the first experiments. It's less than the $3.5 billion cost of the National Ignition Facility, which has yet to achieve true break-even with its laser-blaster fusion experiment. But it's quite a bit more than the unorthodox Polywell fusion reactors proposed by EMC2 Fusion, which are projected to cost in the range of $30 million to $200 million.
Want Safer Nuclear Power? Fusion May Pave the WayMarch 12, 201401:34
One of the researchers on the University of Washington team behind the dynomak design, Derek Sutherland, said the concept could bridge the gap between the expensive mainstream approaches to fusion energy and the low-cost but highly speculative outside-the-box approaches.
"We're like the mainstream enough to benefit from their superior physics, but we're different enough to address the economic issues facing fusion in general," he told NBC News. "Not too alternative, not too mainstream. Maybe it's just right."
On one level, fusion energy is totally mainstream: After all, it's the reaction that powers the sun, which showers our planet with energy from 93 million miles away. It's also the reaction behind hydrogen bombs. In both cases, hydrogen nuclei are smashed together under high pressure and temperature to create helium nuclei. Some of the nuclear mass is converted into pure energy, in accordance with Albert Einstein's E=mc² formula — and that's what produces fusion power.
How much will fusion cost?
Harnessing that reaction has proven devilishly difficult. Decades of work and billions of dollars in spending haven't yet resulted in a reactor that goes beyond the break-even point and produces more energy than it consumes.
Scientists expect ITER to do it in a seven-story-tall facility, but the construction cost is daunting — and even then, the facility will be more of a science experiment than a model for commercial power generation.
"Investors don't want to spend tens of billions of dollars up front, not knowing if the experiment is going to work," Sutherland said.
He explained that the dynomak design is a low-cost cousin of ITER's tokamak design. In both cases, the power-producing plasma is contained inside a magnetic field. But Sutherland said the dynomak is constructed so that "all of the magnetic fields required for stability are generated purely from the currents" in the plasma, rather than from superconducting coils placed around the reactor.
"That enables a more compact reactor," he said.
The University of Washington team, led by physicist Thomas Jarboe, has generated tens of electron volts' worth of fusion power in a small-scale reactor known as HIT-Si3, thanks to U.S. Department of Energy funding. Now the researchers are looking for $8 million to $10 million to build a test reactor that's roughly twice as big. If the computer models are correct, that reactor, dubbed HIT-SiX, would produce temperatures in the range of hundreds of electron volts.
"HIT-SiX will serve as the key risk-reduction experiment," Sutherland said. It wouldn't come anywhere near the break-even point, but it would provide additional data points to show whether the team is really on the right track.
Sutherland said "there's no reason at the moment to believe that the reaction won't scale up to commercial operations." But will reality obey the computer models?
"That's always the million-dollar question," he said. It's also the $8 million, $30 million, $3.5 billion and $50 billion question.
Elsewhere on the fusion frontier:
- A different team of researchers from the University of Washington has been working on yet another approach to fusion power, and some members of that original team have started up a commercial venture called Helion Energy. Helion recently received venture backing from Mithril Capital and YCombinator, and it's made technical progress as well. "We have increased our demonstrated plasma temperatures to over 5 KeV [5,000 electron volts] and continue work on the engineering hardware of our next, break-even machine," Helion CEO David Kirtley told NBC News in an email. Kirtley told The Wall Street Journal that break-even could come in three years, and Xconomy.com quoted him as saying the company could be generating electricity and earning revenue after six years.
- Researchers at Sandia National Laboratories in New Mexico have reported "significant results" from their Z Machine, which crushes hydrogen fuel with colossal pulses of electrical power to spark fusion reactions. Recent experiments, detailed in Physical Review Letters, produced plasma temperatures of about 35 million degrees Celsius and a flux of about 2 trillion neutrons per shot. For more, check out this report from Science's Dalnel Clery.
- An outside review of experiments involving the controversial E-Cat reactor reports evidence of "anomalous heat production." That has raised hopes among E-Cat's fans that cold-fusion technology — also known as low-energy nuclear reaction, or LENR — just might work. Critics have said the heat production is probably due to less exotic reactions that haven't been properly accounted for, but this week's report sides with the view that something nuclear is going on. For more, check out this report from ExtremeTech's Sebastian Anthony.